Tag: Machine Learning

A combined DFT and machine learning study maps how the carbon–fluorine bond in tetrafluoromethane breaks across transition metal surfaces, single-atom ...

A Frontiers editorial outlines how EEG-based biomarkers, machine learning and BCI technologies could support personalised mental healthcare while facing low ...

A two-site EEG study of 9751 participants found reduced alpha power and increased gamma-band connectivity shared across schizophrenia, bipolar disorder ...

A machine-learning radiogenomic model predicts IL19 expression from glioblastoma MRI scans and separates patients into survival groups.

A controlled audit of conformal prediction on two intrusion detection benchmarks shows that marginal coverage can hide rare-class failures as ...

Modeling text embeddings as cyclostationary sequences improves BERT-based multilingual sentiment analysis accuracy by up to 28% across five diverse datasets.

A systematic review of 65 studies finds AI models for PCOS ultrasound diagnosis report high accuracy but suffer from substantial ...

A five-gene machine-learning signature predicts ovarian cancer survival, and the core gene ACSM3 acts as a tumor suppressor by limiting ...

A two-gene transcriptomic signature combining JAK1 and GATA3 reproducibly distinguished sepsis from controls across independent cohorts while tracking shifts from ...

A newly published study shows that StatMatch, a statistical matching-based undersampling method, can match SMOTE on some imbalanced datasets although ...

Researchers developed personalized machine learning algorithms using smart ring biometric data to predict weekly panic attack risks among college students.

A new review concludes that no single DNA marker or marker combination currently works as a universal barcode across bamboo, ...

Researchers report the first systematic feature selection strategies for MinHash-based classifiers, achieving statistically significant gains in most of five real-world ...

Researchers developed and externally validated the DAP5 composite metabolic index to predict diabetes and arthritis comorbidity across cohorts in China ...

A bioinformatics and laboratory study identified FKBP5, MYH6 and PROS1 as candidate endoplasmic reticulum stress-related feature genes in hypertrophic cardiomyopathy.

A computational analysis of 10,509 multilingual news articles found non-EU media used more politically charged framing of social media bans ...

A four-habitat MRI radiomics model combined with clinical variables predicted lymph node metastasis in early-stage cervical cancer with a test ...

Long-range OCT imaging with machine learning segmentation shows that human accommodation is driven mainly by the lens nucleus and innermost ...

A machine-learning framework that explicitly encodes the EPA, WLTP, or NEDC test protocol behind each official electric vehicle range label ...

A new editorial argues that reproducible brain-behavior mappings, not more case-control differences, are the missing link for precision therapeutics in ...